RESUMO
OBJECTIVES: This study evaluated the clinical impact of the Sim&Size® simulation software on the endovascular treatment with flow-diverter stents of patients with unruptured saccular intracranial aneurysms. METHODS: This monocentric retrospective study evaluated a cohort of patients treated with flow-divert stents between June 1, 2014, and December 31, 2019, for cerebral aneurysms. Patients belonged to two groups, patients treated with and without the Sim&Size® simulation software. Univariate, bivariate, and multivariate analyses were used to evaluate the clinical impact of simulation software. RESULTS: Out of the 73 interventions involving 68 patients analyzed by the study, 76.7% were simulated using the Sim&Size® simulation software, and 23.3% were not. Patients treated with the simulation software had shorter stent lengths (16.00â mm vs. 20.00â mm p-value = 0.001) and surgical time (100.00â min vs. 118.00â min p-value = 0.496). Also, fewer of them required more than one stent (3.6% vs. 17.6% p-value = 0.079). Three patients belonging to the non-stimulated group presented hemorrhagic complications. CONCLUSIONS: Using the Sim&Size® simulation software for the endovascular treatment of intracranial aneurysms with pipeline flow-diverter stents reduces the stent length.
Assuntos
Embolização Terapêutica , Procedimentos Endovasculares , Aneurisma Intracraniano , Humanos , Aneurisma Intracraniano/terapia , Estudos Retrospectivos , Resultado do Tratamento , Embolização Terapêutica/métodos , Stents/efeitos adversos , Software , Procedimentos Endovasculares/métodos , Angiografia CerebralRESUMO
BACKGROUND: Thromboembolic events are critical complications in neuroendovascular procedures, and dual antiplatelet therapy (DAPT) can reduce them. The effects of using aspirin and clopidogrel in DAPT are well characterized, but use of aspirin and ticagrelor has been less studied. METHODS: This retrospective cohort study, conducted between April 1, 2015, and December 30, 2020, included patients with endovascular treatment with flow-diverting and non-flow-diverting stents for unruptured cerebral aneurysms who received DAPT with aspirin and clopidogrel or with aspirin and ticagrelor. RESULTS: Of 148 patients with unruptured intracranial aneurysms with flow-diverting and non-flow-diverting stents started on DAPT with aspirin (100 mg/day) and clopidogrel (75 mg/day), 24 had a poor response to clopidogrel according to the VerifyNow test and had DAPT changed to aspirin (100 mg/day) and ticagrelor (90 mg every 12 hours). One thrombotic complication (0.81%) and 1 bleeding complication (0.81%) occurred in patients receiving DAPT with clopidogrel and aspirin during the procedure. These complications did not occur (0.00%) in patients receiving DAPT with ticagrelor and aspirin. At the 6-month follow-up, 4 patients (3.15%) in the clopidogrel group presented with thrombotic complications, whereas no patients (0.00%) in the ticagrelor group experienced this complication. At 6-month follow-up, 4 patients (3.23%) in the clopidogrel group presented with hemorrhagic complications, whereas only 1 patient (4.17%) in the ticagrelor group experienced this complication. CONCLUSIONS: Our study showed that DAPT with ticagrelor (90 mg every 12 hours) and aspirin (100 mg/day) is a safe and effective alternative to DAPT with clopidogrel (75 mg/day) and aspirin (100 mg/day) for patients with an inadequate response to clopidogrel.
RESUMO
INTRODUCTION: The coronavirus disease 2019 (COVID-19) has become a significant public health problem worldwide. In this context, CT-scan automatic analysis has emerged as a COVID-19 complementary diagnosis tool allowing for radiological finding characterization, patient categorization, and disease follow-up. However, this analysis depends on the radiologist's expertise, which may result in subjective evaluations. OBJECTIVE: To explore deep learning representations, trained from thoracic CT-slices, to automatically distinguish COVID-19 disease from control samples. MATERIALS AND METHODS: Two datasets were used: SARS-CoV-2 CT Scan (Set-1) and FOSCAL clinic's dataset (Set-2). The deep representations took advantage of supervised learning models previously trained on the natural image domain, which were adjusted following a transfer learning scheme. The deep classification was carried out: (a) via an end-to-end deep learning approach and (b) via random forest and support vector machine classifiers by feeding the deep representation embedding vectors into these classifiers. RESULTS: The end-to-end classification achieved an average accuracy of 92.33% (89.70% precision) for Set-1 and 96.99% (96.62% precision) for Set-2. The deep feature embedding with a support vector machine achieved an average accuracy of 91.40% (95.77% precision) and 96.00% (94.74% precision) for Set-1 and Set-2, respectively. CONCLUSION: Deep representations have achieved outstanding performance in the identification of COVID-19 cases on CT scans demonstrating good characterization of the COVID-19 radiological patterns. These representations could potentially support the COVID-19 diagnosis in clinical settings.
Introducción. La enfermedad por coronavirus (COVID-19) es actualmente el principal problema de salud pública en el mundo. En este contexto, el análisis automático de tomografías computarizadas (TC) surge como una herramienta diagnóstica complementaria que permite caracterizar hallazgos radiológicos, y categorizar y hacer el seguimiento de pacientes con COVID-19. Sin embargo, este análisis depende de la experiencia de los radiólogos, por lo que las valoraciones pueden ser subjetivas. Objetivo. Explorar representaciones de aprendizaje profundo entrenadas con cortes de TC torácica para diferenciar automáticamente entre los casos de COVID-19 y personas no infectadas. Materiales y métodos. Se usaron dos conjuntos de datos de TC: de SARS-CoV-2 CT (conjunto 1) y de la clínica FOSCAL (conjunto 2). Los modelos de aprendizaje supervisados y previamente entrenados en imágenes naturales, se ajustaron usando aprendizaje por transferencia. La clasificación se llevó a cabo mediante aprendizaje de extremo a extremo y clasificadores tales como los árboles de decisiones y las máquinas de soporte vectorial, alimentados por la representación profunda previamente aprendida. Resultados. El enfoque de extremo a extremo alcanzó una exactitud promedio de 92,33 % (89,70 % de precisión) para el conjunto 1 y de 96,99 % (96,62 % de precisión) para el conjunto-2. La máquina de soporte vectorial alcanzó una exactitud promedio de 91,40 % (precisión del 95,77 %) para el conjunto-1 y del 96,00 % (precisión del 94,74 %) para el conjunto 2. Conclusión. Las representaciones profundas lograron resultados sobresalientes al caracterizar patrones radiológicos usados en la detección de casos de COVID-19 a partir de estudios de TC y demostraron ser una potencial herramienta de apoyo del diagnóstico.
Assuntos
COVID-19 , Aprendizado Profundo , Teste para COVID-19 , Humanos , Redes Neurais de Computação , SARS-CoV-2 , Tomografia Computadorizada por Raios XRESUMO
Introduction: The coronavirus disease 2019 (COVID-19) has become a significant public health problem worldwide. In this context, CT-scan automatic analysis has emerged as a COVID-19 complementary diagnosis tool allowing for radiological finding characterization, patient categorization, and disease follow-up. However, this analysis depends on the radiologist's expertise, which may result in subjective evaluations. Objective: To explore deep learning representations, trained from thoracic CT-slices, to automatically distinguish COVID-19 disease from control samples. Materials and methods: Two datasets were used: SARS-CoV-2 CT Scan (Set-1) and FOSCAL clinic's dataset (Set-2). The deep representations took advantage of supervised learning models previously trained on the natural image domain, which were adjusted following a transfer learning scheme. The deep classification was carried out: (a) via an end-to-end deep learning approach and (b) via random forest and support vector machine classifiers by feeding the deep representation embedding vectors into these classifiers. Results: The end-to-end classification achieved an average accuracy of 92.33% (89.70% precision) for Set-1 and 96.99% (96.62% precision) for Set-2. The deep feature embedding with a support vector machine achieved an average accuracy of 91.40% (95.77% precision) and 96.00% (94.74% precision) for Set-1 and Set-2, respectively. Conclusion: Deep representations have achieved outstanding performance in the identification of COVID-19 cases on CT scans demonstrating good characterization of the COVID-19 radiological patterns. These representations could potentially support the COVID-19 diagnosis in clinical settings.
Introducción. La enfermedad por coronavirus (COVID-19) es actualmente el principal problema de salud pública en el mundo. En este contexto, el análisis automático de tomografías computarizadas (TC) surge como una herramienta diagnóstica complementaria que permite caracterizar hallazgos radiológicos, y categorizar y hacer el seguimiento de pacientes con COVID-19. Sin embargo, este análisis depende de la experiencia de los radiólogos, por lo que las valoraciones pueden ser subjetivas. Objetivo. Explorar representaciones de aprendizaje profundo entrenadas con cortes de TC torácica para diferenciar automáticamente entre los casos de COVID-19 y personas no infectadas. Materiales y métodos. Se usaron dos conjuntos de datos de TC: de SARS-CoV-2 CT (conjunto 1) y de la clínica FOSCAL (conjunto 2). Los modelos de aprendizaje supervisados y previamente entrenados en imágenes naturales, se ajustaron usando aprendizaje por transferencia. La clasificación se llevó a cabo mediante aprendizaje de extremo a extremo y clasificadores tales como los árboles de decisiones y las máquinas de soporte vectorial, alimentados por la representación profunda previamente aprendida. Resultados. El enfoque de extremo a extremo alcanzó una exactitud promedio de 92,33 % (89,70 % de precisión) para el conjunto 1 y de 96,99 % (96,62 % de precisión) para el conjunto-2. La máquina de soporte vectorial alcanzó una exactitud promedio de 91,40 % (precisión del 95,77 %) para el conjunto-1 y del 96,00 % (precisión del 94,74 %) para el conjunto 2. Conclusión. Las representaciones profundas lograron resultados sobresalientes al caracterizar patrones radiológicos usados en la detección de casos de COVID-19 a partir de estudios de TC y demostraron ser una potencial herramienta de apoyo del diagnóstico.
Assuntos
Infecções por Coronavirus/diagnóstico , Aprendizado Profundo , Tomografia Computadorizada por Raios XRESUMO
Introducción: Las endofugas son la complicación más frecuente de los tratamientos endovasculares de aneurismas de aorta abdominal y torácica. El objetivo de este estudio es describir la frecuencia de endofugas en pacientes con aneurismas de aorta infrarrenal tratados con técnicas endovasculares. Metodología: Estudio de cohorte retrospectivo en el que se incluyeron pacientes con aneurismas infrarrenales tratados con terapia endovascular en dos instituciones de alta complejidad entre el 1 de septiembre de 2013 y el 1 de marzo de 2021. Se incluyeron datos demográficos, antecedentes, características morfológicas del cuello y saco del aneurisma, tipo de prótesis utilizada, presencia y tipo de endofuga. Se realizó un análisis descriptivo univariado. Los intervalos de confianza se describieron con un 95%. Resultados: Se incluyeron 99 pacientes, la media de edad fue 74,37 años, la media de la longitud del cuello fue de 29,47 mm, el 90,24% tuvieron una longitud favorable (>15mm). La media del ángulo fue de 44,57°, el 67,86% tenía un ángulo favorable (<60°). El 28,28% de los pacientes presentaron endofugas, la frecuencia de las endofugas tipo Ia fue de 7,07%, las de tipo Ib 8,08%, las de tipo II 18,37%, las de tipo IIIa y IIIb 1,01%. No se presentaron endofugas tipo IV ni V. Conclusiones: La frecuencia de presentación de endofugas fue del 28,28%; la endofuga más frecuente es la de tipo II 18,37%, ligeramente inferior a lo descrito en la literatura.
Introduction: Endoleaks are the most common complication of endovascular treatment of abdominal and thoracic aortic aneurysms.. The objective of this study is to describe the frequency of endoleaks in patients with infrarenal aortic aneurysms treated with endovascular techniques. Methodology: Retrospective cohort study that included patients from September 1, 2013, to March 1, 2021, with infrarenal aneurysms treated with endovascular therapy at the FOSCAL and FOSCAL international clinics. Demographic data, history, morphological characteristics of the aneurysm neck and sac, type of prosthesis used, presence, and type of endoleak were included. A univariate descriptive analysis was performed. Confidence intervals were reported at 95%. Results: 99 patients were included, the mean age was 74.37 years, the mean neck length was 29.47 mm, 90.24% had a favorable length (>15 mm); The mean angle was 44.57, 67.86% had a favorable angle (<60º). 28.28% of the patients presented endoleaks, the frequency of type Ia endoleaks was 7.07%, type Ib endoleaks 8.08%, type II 18.37%, type IIIa, and IIIb endoleaks 1, 01%. There were no type IV or type V endoleaks. Conclusions: The frequency of presentation of endoleaks was 28.28%; the most frequent endoleak is type II 18.37%. slightly lower than that reported in the literature